- Email: [email protected]
Corrosion of cold-deformed brass in acid sulphate solution
Corrosion Science 46 (2004) 2793–2802 www.elsevier.com/locate/corsci
Corrosion of cold-deformed brass in acid sulphate solution Zoran Avramovic
a,*
, Milan Antonijevic
b
a
b
Copper Institute, Zeleni bulevar 33–35, 19210 Bor, Yugoslavia (Serbia) University in Belgrade, Technical Faculty Bor, VJ 12, 19210 Bor, Yugoslavia (Serbia) Received 8 July 2002; accepted 19 March 2004 Available online 10 May 2004
Abstract The corrosion behaviour and the dezincification process of cold-deformed CuZn-42 brass were tested in an acid sulphate solution at pH-value 2 with additional chloride and copper(II) ions by use of the linear polarization method. The measured corrosion potential and densities of corrosion currents were observed as characteristics of the dezincification process and the corrosion resistance of tested samples of cold-deformed CuZn-42 brass. The results obtained have shown that pH-value 2 of the tested solutions and increased concentrations of copper(II) ions result in increased values of densities of corrosion currents of the tested brass samples, as a result of selective zinc dissolution and the individual dissolution of zinc and copper including the process of anodic oxidation. The tested concentrations of chloride ions in certain conditions have an inhibiting effect, whereas in the other conditions they act as distinctive activators of brass corrosion. The lowest values of corrosion currents are present in the brass samples with the highest deformation degree at 80%. The process of dezincification and anode dissolution of cold deformed brass samples were developed in the whole range of tested potentials. 2004 Elsevier Ltd. All rights reserved. Keywords: Brass; Polarization; Dealloying; Deformation degrees; Oxidation
1. Introduction The corrosion behaviour of brass has been studied from an aspect of dezincification mechanism, dealloying, and stress corrosion. Dezincification is a well-known *
Corresponding author. Tel.: +381-30-432-299x731; fax: +381-30-436-728. E-mail address: [email protected] (Z. Avramovic).
0010-938X/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.corsci.2004.03.010
2794
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
dealloying process that designates a loss of brass as an important physical/ mechanical characteristic resulting in surface destruction [1–7]. It appears in solutions with a content of specific chemical compounds [1,3–5]. Previous works have established that ions as thiozionates, bromides, and iodides do not participate in dezincification whereas chlorides and sulphates stimulate dezincification [1–5]. The influence of anions was reflected in a distinctive effect on the sample surface. The noticed differences in dissolution kinetics and dezincification degree in chloride and sulphate media could have been caused by a connection between copper and the existing anions. Two basic theories explain the dezincification mechanism of brass as well: One theory assumes a selective zinc dissolution, which is released from alloy, and the porous residue reached with metal copper is lagged, and, the other theory assumes a simultaneous dissolution of zinc and copper where copper redeposition is developed at a suitable degree. Some opposing opinions are still held about copper redeposition as well as the form and character of brass corrosion products that are separated on the sample surface during selective alloy dissolution [8–11]. The inhibiting effect of chloride ions on brass corrosion, shown by many authors [8–10,12,13] in neutral ammonium solution, have chloride ions concentration of 8 · 103 M. Later, Uhlig and co-authors [10] investigated the inhibiting effect of Cl ions on corrosion 63Cu–37Zn brass and found out the optimum concentration of Cl ions for corrosion inhibition is 4 · 102 N. Tests carried out in a 101 M solution of Na2 SO4 , with addition of various concentrations the Cu2þ and Cl ions, showed that the highest corrosion of tested brass was at a concentration of chloride ions of 5 · 103 N whereas the inhibiting effect presented at higher concentrations of Cl ions. Anode polarization potentiometric-dynamic measurements were also carried out in a 101 M solution of Na2 SO4 , with a varied content of Cl ions at pH ¼ 1.5–3.5 [9,10]. At constant acidity, the polarization curves showed an insignificant influence of Cl ions when an added concentration of Cl ions is lower than 1 · 102 N, but, the polarization curves showed the presence of anode inhibition at higher concentrations of Cl ions. CuCl and Cu2 O films are developed in solutions with a pH-value of 3–3.5, with a presence of chloride ions in concentrations of 103 –102 N and of copper sulphate in concentrations of 5 · 102 M. An insignificant effect on corrosion was present in the same solution with concentrations of chloride ions lower than 103 N, explained by the fact that the solid stage is spread in a Cu2 O film in such cases whereas the inhibiting effect presented at higher concentrations of Cl ions (l M). This was attributed to the formation of a stable CuCl deposit. A Cu2 O film was formed from growths in the copper sulphate solution, that are in and epotexial connection with the base. It enables a deposit growth on a deformed base that could have various growths and become more prevalent to the local dissolution [13]. According to G. Bianchin, the dezincification process was developed with help of a high transfer degree of mass that inhibited a protective film formation of corrosive products, especially the Cu2 O film [14]. The effect of Cl ions in solutions on copper oxidation acceleration was in relation to the defect of changed Cu2 O-film structure
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2795
that had a lower protection degree on it. It assumed that the Cl ion from solution replaced some O2 ions in the oxide grid, where the primary process of selective brass dissolution (zinc dissolution) was expressed by a sudden increase of the current density, short over corrosion potential. The brass dezincification process in solution with a content of chloride ions was developed by a selective dissolution of a more reactive component (zinc), where surface diffusion of a more stable component (copper) was added. A. F. Lucey has pointed out the importance of the formation the Cu(I)-complex ion in a NaCl solution for the process of selective brass dissolution [14]. Based on corrosion testing of the copper-based alloys in sea water, de Sanchez and Schiffrin [14,15] have assumed that the formation of protective films depends a on diffusion balance of Cu(I)-chloro-complex that is formed by a primary anode reaction: Cu þ 2Cl ! CuCl 2 þe
ð1Þ
and by deposition reaction: 2CuCl 2 þ 2OH ! Cu2 O þ H2 O þ 4Cl
ð2Þ
A distinct dissolubility of Cu(I)-chloro-complex, in high salinity solutions, sets up the conditions for the development process of selective brass dissolution in the sea. In this work, the authors have tried to explain the corrosion behaviour of samples from cold-deformed CuZn-42 brass in a function of concentration with the chloride and copper(II) ions, pH-values of solutions, and the deformation degree.
2. Experimental part The samples of tested brass that were laboratory produced had the following chemical content: Cu 57.95% (99.997% purity), Zn 41.91% (99.85% purity), and others 0.14%. Samples were deformed to the following deformation degrees: 20%, 40%, 60% and 80% and sealed in cold-polymerized acrylate. Deformations were carried out in laboratory conditions by the use of cold rolling the samples between two parallel rollers, where the deformation was extended over the whole length of the sample height in all directions. A non-deformed copper electrode (99.999% purity) was used as a comparative sample. The samples for the electrochemical measurements had a constant area of P ¼ 0:38 cm2 . Before every polarization measurement, the samples were polished on emery paper (fineness #1000) and alumina rinsed with distilled water and ethyl alcohol. A saturated calomel electrode (SCE) and platinum wire were used as a reference, i.e. counter electrode, separately, in a classic threeelectrode electrochemical cell. The all potential values were given regarding the saturated calomel electrode. Tested solutions were: 101 M Na2 SO4 , 101 M Na2 SO4 + 1 · 103 M CuSO4 , 101 M Na2 SO4 + 5 · 103 M CuSO4 , 101 M Na2 SO4 + 1 · 102 M CuSO4 , 101 M Na2 SO4 + 5 · 102 M CuSO4 , 101 M Na2 SO4 + 5 · 104 M NaCl, 101 M Na2 SO4 + 5 · 103 M NaCl, 101 M Na2 SO4 + 5 · 102 M NaCl, 101 M
2796
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
Na2 SO4 + 101 M NaCl, 101 M Na2 SO4 + 5 · 101 M NaCl, 101 M Na2 SO4 + 1.0 M NaCl and preparations of high pure chemicals with distilled water. All experiments have been done with a solution volume of 50 cm3 . The pH-value 2 of solutions was adjusted with a 101 M solution of H2 SO4 . The operating temperature was 20 C, with the use of open atmosphere. Polarization measurements were carried out from a potential of open circuit to a potential of 1100 mV (SCE), with the polarization rate of 10 mV/s, by use of the linear polarization method. Corrosion current densities were determined by approximating the right part of anode Tafel’s curves to the section with the corrosion potential. The potential region from 0 (vs. SCE) to 200 mV (vs. SCE) was taken for the calculation area. An AMEL apparatus was used as well as: potentiostat (Model 553), program functional generator (Model 568), interface (Model 560/A/ log), and digital x/y-recorder (Model 863).
3. Experimental results and discussion 3.1. Influence of copper(II) ions Figs. 1 and 2 present the dependence of current density from the potential for tested copper and brass samples in an 101 M solution of Na2 SO4 with the addition of 5 · 102 M Cu2þ ions at pH 2. Corrosion potentials were established upon 15–20 min intervals, where the potential for copper electrode was shifted to more negative values, and, for brass electrode to more positive values with time. In the 101 M solution of Na2 SO4 with and without the addition of Cu2þ ions, the values of corrosion potentials for brass electrodes were approximate and lower than the values for copper electrode where a small pH-value of the solution shifted values of corrosion potential in the more negative region, that could be explained by local reactions: 1000 Cu (Ecorr=-30mV)(jcorr=0.511mA/cm2) CuZn-42(0%),(-88mV)(0.681) CuZn-42(20%),(-85mV)(0.738) CuZn-42(40%),(-85mV)(0.767) CuZn-42(60%),(-88mV)(0.912) CuZn-42(80%),(-84mV)(0.668)
E (mV/SCE)
800
600
0.1M Na2SO4 pH=2
400
200
0 2.5
3.0
3.5
4.0
4.5
5.0
log(j, µA/cm2)
Fig. 1. Polarization curves for tested samples in 101 M Na2 SO4 , at pH 2.
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2797
1000 C u , ( E corr= 3 1 m V ) ( j corr= 0 . 9 4 7 m A / c m 2) CuZn-42(0%),(-16mV)(1.238) CuZn-42(20%),(10mV)(1.333) CuZn-42(40%),(10mV)(1.378) CuZn-42(60%),(-10mV)(1.448) CuZn-42(80%),(10mV)(1.211)
E (mV/SCE)
800
600
0.1M Na2SO4 + 5*10-2M Cu2+ pH=2
400
200
0 2.8
3.2
3.6
4.0
4.4
4.8
5.2
log( j, µA/cm2)
Fig. 2. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 102 M Cu2þ , at pH 2.
Zn ! Zn2þ þ 2e
ð3Þ
Cu2þ þ 2e ! Cu
ð4Þ
on sample surface [16]. The concentration increase of Cu2þ ions resulted in the movement of corrosion potentials into a less active region, pointing out that the balance realization between Cu2þ /Cuþ ions on a boundary of metal–solution and the values of current densities are also increased [13]. The lowest values of corrosion current densities (jcorr ) were for brass with a deformation degree of 80%. The dezincification process, present here, is a cause of the increase of the density values of corrosive currents as well as the non-existence of a protective film that mainly consisted of Cu2 O on a sample surface at pH 2. At higher pH-values, protective film, besides Cu2 O and CuO oxides, also consisted of a Cu3 (SO4 )2 Æ 4H2 O component that was stable in a wide range of pH-values [3,17]. The corrosive morphology of cold-deformed samples has a disadvantage in acid solutions because it has a resistant protective film only to a certain degree, under conditions of local destruction (dissolution) and gives corrosive openings that a little bit later resulted in spreading of one type of corrosion [13]. Cuþ ion, present in Cuþ compounds, probably is an intermediate form in a reaction of dissolution followed by a controlled oxidation rate of Cuþ to Cu2þ ions, where more favourable conditions have been provided for brass dezincification [3,14,17]. 3.2. Influence of chloride ions The polarization curves in Figs. 3 and 4 are shown for copper and brass with various deformation degrees in a 101 M Na2 SO4 solution with the addition of chloride ions in a concentration of 5 · 104 and 5 · 103 M at pH 2. Immediately
2798
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
C u , ( E corr= - 5 5 m V ) ( j corr= 0 . 7 2 7 m A / c m 2) CuZn-42(0%),(-149mV)(1.04) CuZn-42(20%),(-171mV)(1.108) CuZn-42(40%),(-136mV)(1.166) CuZn-42(60%),(-148mV)(1.42) CuZn-42(80%),(-170mV)(1.035)
1000
E (mV/SCE)
800
600
0.1M Na2SO4 + 5*10-4 ClpH=2
400
200
0 2.0
2.5
3.0
3.5
4.0
4.5
5.0
log( j, µA/cm2)
Fig. 3. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 104 M Cl , at pH 2.
1200 C u ( E corr= - 6 1 m V ) , ( j corr= 0 . 9 8 8 m A / c m 2) CuZn-42(0%),(-195mV),(1.298) CuZn-42(20%),(-188mV),(1.585) CuZn-42(40%),(-160mV),(1.688) CuZn-42(60%),(-170mV),(1.721) CuZn-42(80%),(-220mV),(1.278)
1000
E (mV/SCE)
800 600
-
0.1M Na2SO4 + 5*10-3M Cl pH=2
400 200 0 -200 2.5
3.0
3.5
4.0
4.5
5.0
log( j, µA/cm2)
Fig. 4. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 103 M Cl , at pH 2.
upon the setting of corrosion potential, all tested samples are moved into an active state where they are in the whole range of tested potential. With an increased Cl ion concentration, the corrosion potentials have a tendency of movement into a more active region for all the tested electrodes, inducing a lower degree of anode process control. The most positive values for Ecorr in the Cu electrode are in all tested solutions. The anode polarization curves have a similar form unless the values for jcorr are also increased with an increased brass deformation degree up to 60%, whereas values of the corrosion current densities are the lowest for a deformation degree of 80%. Only copper electrode has lower values for jcorr of brass with a deformation degree of 80%.
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2799
Densities of corrosion currents are higher than solutions with a concentration of chloride ions of 5 · 103 M. It could be noted here that the values for jcorr with a 101 M solution of Na2 SO4 and an addition of chloride ions in a concentration of 5 · 103 M are higher than values for jcorr with a solution concentration of Cl ions of 5 · 104 and 5 · 102 M, even higher than some values with a solution concentration of chloride ions of 101 M. Figs. 5 and 6 give anode polarization curves for tested brass samples in a 101 M solution of Na2 SO4 with an addition of chloride ions in a concentration of 5 · 102 and 1.0 M. The values of corrosive potentials relevant to an 101 M solution of Na2 SO4 , with an addition of chloride ions in concentration of 5 · 104 M, were shifted into a more active region. The values of corrosion current densities were moved into a more positive area at small pH-values of solution and an increase of brass deformation
C u , ( E corr= - 1 4 5 m V ) ( j corr= 0 . 6 7 8 m A / c m 2) CuZn-42(0%),(-232mV)(1.083) CuZn-42(20%),(-226mV)(1.122) CuZn-42(40%),(-240mV)(1.182) CuZn-42(60%),(-218mV)(1.192) CuZn-42(80%),(-225mV)(1.008)
1000
E (mV/SCE)
800 600
0.1M Na2SO4 + 5*10-2M Cl pH=2
-
400 200 0 -200 2.5
3.0
3.5
4.0
4.5
5.0
log( j, µA/cm2)
Fig. 5. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 102 M Cl , at pH 2.
C u , ( E corr= - 2 7 3 m V ) , ( j corr= 1 . 4 4 2 m A / c m 2) CuZn-42(0%),(-335mV)(3.158) CuZn-42(20%),(-330mV)(3.492) CuZn-42(40%),(-330mV)(3.752) CuZn-42(60%),(-330mV)(4.11) CuZn-42(80%),(-333mV)(3.005)
1000
E (mV/SCE)
800 600
-
400
0.1M Na2SO4 + 1.0M Cl pH=2
200 0 -200 2.5
3.0
3.5
4.0
4.5
5.0
5.5
log(j, µA/cm2)
Fig. 6. Polarization curves for tested samples in 101 M Na2 SO4 + 1.0 M Cl , at pH 2.
2800
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
degree up to 60%, where values for jcorr are the highest. The most negative values for jcorr are in brass with a deformation degree of 80%. On anode polarization curves, the presence of anode peaks were noticed, presenting an active dissolution of tested samples and formation of Cu(I)-chloro complex formed by a primary reaction: Cu + 2Cl $ CuCl 2 + e . The Cu(I)-chloro 2 complex that will be formed (CuCl2 or CuCl3 ) depends on the concentration of chloride ions. A distinct dissolution of Cu(I)-chloro-complex, in solutions with an increased concentration of Cl ions, makes conditions for development of the selective dissolution process of brass (dezincification) as well as for a loss of the already formed copper-chloride film on its surface. In concentration values of chloride ions of 5 · 102 M, a drop of value for jcorr was noticed regarding to the solutions with an addition of chloride ions in a concentration of 5 · 103 M, explained by an inhibitory effect of higher concentrations of chloride ions (see Fig. 7). Anode polarization curves show that the lowest values for corrosive current densities are in Cu electrodes; and, from brass samples, that brass with a deformation degree of 80% has the lowest values for jcorr . Brass with a deformation of 60% has the highest values for jcorr . Regarding the Cu electrode, the anode peaks of brass electrodes are moved to more positive potential values. Flow of anode polarization curves pointed out that anode dissolution of brass in solutions with chloride ions started with an initial primary dissolution of zinc, and then zinc and copper, depending on potential region, but not in a stoichiometric relationship. The primary process of selective brass dissolution was expressed by a sudden increase of current density immediately upon the corrosion potential, at the beginning of anode polarization. By chemical analysis of the analyzed solution, presence of some Zn2þ and Cu2þ /Cuþ ions was found to point out that, under the afore mentioned experimental condition, the dezincification process of brass could be developed with the following process of its anode dissolution. Various electrode potentials of zinc, that is copper, have considerable influence on the degree, intensity, and rate of the pri-
4.0 3.5
CuZn-42 (60%) pH-2 0.1M Na2SO4
jcorr (mA/cm2)
3.0 2.5 2.0 1.5 1.0 0.0
0.2
0.4 0.6 CCl- (M)
0.8
1.0
Fig. 7. Dependence of jcorr on concentration of Cl ions for CuZn-42 (60%), at pH 2.
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2801
mary anode dissolution. Zinc passage into a solution in the form of Zn2þ ion and the formation of complex and undissolved salts contributes to the irreversible nature of development in the dezincification process under the above mentioned conditions. Metallic copper drops back, partly as a porous metallic mass on an electrode surface of CuZn-42 brass, whereas the other part of Cu ions diffuses on a brass sample surface, where it is separated again in a form of copper and various Cu compounds by further kinetics control of total alloy dissolution. According to the balance diagram for content of Cu–CuCl , at various pH-values of NaCl solution, it was found that the major participation of more dissoluble 2 copper compounds (Cuþ , CuCl 2 , CuCl3 ) are present at higher pH-values relevant to the participation of less dissoluble copper compounds (CuCl, Cu2 O) [14]. Based on the obtained results, the brass dezincification process is developed, in solutions with a content of chloride ions, by the selective dissolution of a more reactive component (zinc) where surface diffusion of more a stable component (copper) has been added to. The dezincification process was developed by help of a high degree of mass transfer that inhibited the protective film formation of corrosive products, especially Cu2 O film [8,15,16]. Lesser values of corrosion current densities, in tested samples with a deformation degree of 80% and at higher concentrations of chloride ions, are explained by an easier process of substitution O2 ions with chloride ions in crystal lattice. Also, a protective film is formed faster in deformed samples of 80% on its surface that causes lower values for jcorr .
4. Conclusion Based on the presented testing results of corrosion behaviour of CuZn-42 brass, with various deformation degrees, the following conclusions could be made: 1. Increase of Cu2þ ion concentration results in an increased value for jcorr . 2. Inhibiting effect of chloride ions was noticed for concentrations of ions of 5 · 102 and 101 M in an 101 M solution of Na2 SO4 . 3. Increase of a concentration of the Cl ions, except for values of 5 · 102 and 101 M, resulted in a significant increase of value for jcorr . 4. A pH-value of solution resulted in an increase of value for jcorr . 5. The process of dezincification and anode dissolution of CuZn-42 brass was developed in the whole range of tested potentials, and pointed out the unstability of formed films on their surface. 6. The complex form of anode curves pointed out the phase reactions of selective a dissolution process of CuZn-42 brass. 7. Surface film formation of corrosion products was developed through some phases of and distinctive presence of metallic copper layers and spot corrosion present on surface of corroded electrode. 8. Brass with a deformation degree of 80% has the lowest values for corrosion current densities of brass.
2802
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
Acknowledgements The authors wish to thank the Ministry of Science and Technology of Serbia for financial support (No. 1622).
References [1] R.H. Heidersbach Jr., Ellis D. Verink Jr., Corrosion-NACE 28 (11) (1972) 397. [2] T.K.G. Namboodhiri, R.S. Chaudhary, B. Prakash, M.K. Agrawal, Corrosion Science 22 (1982) 1037. [3] R.K. Dinnappa, S.M. Mayanna, Corrosion Science 27 (1987) 349. [4] R.B. Rebak, J.R. Galvele, Corrosion Science 29 (1989) 1003. [5] L. Burzynska, J. Stoch, Z. Zembura, Solid State Ionics 38 (1990) 179. [6] L. Burzynska, Z. Zembura, Polish Journal of Chemistry 66 (1992) 503. [7] L. Burzynska, A. Maraszewska, Z. Zembura, Corrosion Science 38 (1996) 337. [8] Lj. Krstulovic, B. Kulusic, Chemistry and Industry 45 (5) (1996) 177. [9] T.P. Hoar, C.J.L. Booker, Corrosion Science 5 (1965) 821. [10] S. Torchio, F. Mazza, Corrosion Science 26 (1986) 813. [11] S.S. Birley, Desmond Tromans, Corrosion-NACE 27 (7) (1971) 297. [12] R.C. Newman, G.T. Burstein, Corrosion Science 21 (1981) 119. [13] S. Torchio, Corrosion Science 26 (1986) 133. [14] B. Kulusic, Lj. Krstulovic, J. Ivic, Chemistry and Industry 39 (7) (1990) 315. [15] S.R. De Sanchez, D.J. Schiffrin, Corrosion Science 28 (1988) 141. [16] F. Mansfeld, G. Liu, H. Xiao, C.H. Tsai, B.J. Little, Corrosion Science 36 (12) (1994) 2063. [17] A.M. Beccaria, E.D. Mor, G. Poggi, F. Mazza, Corrosion Science 27 (1987) 363.
Corrosion of cold-deformed brass in acid sulphate solution Zoran Avramovic
a,*
, Milan Antonijevic
b
a
b
Copper Institute, Zeleni bulevar 33–35, 19210 Bor, Yugoslavia (Serbia) University in Belgrade, Technical Faculty Bor, VJ 12, 19210 Bor, Yugoslavia (Serbia) Received 8 July 2002; accepted 19 March 2004 Available online 10 May 2004
Abstract The corrosion behaviour and the dezincification process of cold-deformed CuZn-42 brass were tested in an acid sulphate solution at pH-value 2 with additional chloride and copper(II) ions by use of the linear polarization method. The measured corrosion potential and densities of corrosion currents were observed as characteristics of the dezincification process and the corrosion resistance of tested samples of cold-deformed CuZn-42 brass. The results obtained have shown that pH-value 2 of the tested solutions and increased concentrations of copper(II) ions result in increased values of densities of corrosion currents of the tested brass samples, as a result of selective zinc dissolution and the individual dissolution of zinc and copper including the process of anodic oxidation. The tested concentrations of chloride ions in certain conditions have an inhibiting effect, whereas in the other conditions they act as distinctive activators of brass corrosion. The lowest values of corrosion currents are present in the brass samples with the highest deformation degree at 80%. The process of dezincification and anode dissolution of cold deformed brass samples were developed in the whole range of tested potentials. 2004 Elsevier Ltd. All rights reserved. Keywords: Brass; Polarization; Dealloying; Deformation degrees; Oxidation
1. Introduction The corrosion behaviour of brass has been studied from an aspect of dezincification mechanism, dealloying, and stress corrosion. Dezincification is a well-known *
Corresponding author. Tel.: +381-30-432-299x731; fax: +381-30-436-728. E-mail address: [email protected] (Z. Avramovic).
0010-938X/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.corsci.2004.03.010
2794
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
dealloying process that designates a loss of brass as an important physical/ mechanical characteristic resulting in surface destruction [1–7]. It appears in solutions with a content of specific chemical compounds [1,3–5]. Previous works have established that ions as thiozionates, bromides, and iodides do not participate in dezincification whereas chlorides and sulphates stimulate dezincification [1–5]. The influence of anions was reflected in a distinctive effect on the sample surface. The noticed differences in dissolution kinetics and dezincification degree in chloride and sulphate media could have been caused by a connection between copper and the existing anions. Two basic theories explain the dezincification mechanism of brass as well: One theory assumes a selective zinc dissolution, which is released from alloy, and the porous residue reached with metal copper is lagged, and, the other theory assumes a simultaneous dissolution of zinc and copper where copper redeposition is developed at a suitable degree. Some opposing opinions are still held about copper redeposition as well as the form and character of brass corrosion products that are separated on the sample surface during selective alloy dissolution [8–11]. The inhibiting effect of chloride ions on brass corrosion, shown by many authors [8–10,12,13] in neutral ammonium solution, have chloride ions concentration of 8 · 103 M. Later, Uhlig and co-authors [10] investigated the inhibiting effect of Cl ions on corrosion 63Cu–37Zn brass and found out the optimum concentration of Cl ions for corrosion inhibition is 4 · 102 N. Tests carried out in a 101 M solution of Na2 SO4 , with addition of various concentrations the Cu2þ and Cl ions, showed that the highest corrosion of tested brass was at a concentration of chloride ions of 5 · 103 N whereas the inhibiting effect presented at higher concentrations of Cl ions. Anode polarization potentiometric-dynamic measurements were also carried out in a 101 M solution of Na2 SO4 , with a varied content of Cl ions at pH ¼ 1.5–3.5 [9,10]. At constant acidity, the polarization curves showed an insignificant influence of Cl ions when an added concentration of Cl ions is lower than 1 · 102 N, but, the polarization curves showed the presence of anode inhibition at higher concentrations of Cl ions. CuCl and Cu2 O films are developed in solutions with a pH-value of 3–3.5, with a presence of chloride ions in concentrations of 103 –102 N and of copper sulphate in concentrations of 5 · 102 M. An insignificant effect on corrosion was present in the same solution with concentrations of chloride ions lower than 103 N, explained by the fact that the solid stage is spread in a Cu2 O film in such cases whereas the inhibiting effect presented at higher concentrations of Cl ions (l M). This was attributed to the formation of a stable CuCl deposit. A Cu2 O film was formed from growths in the copper sulphate solution, that are in and epotexial connection with the base. It enables a deposit growth on a deformed base that could have various growths and become more prevalent to the local dissolution [13]. According to G. Bianchin, the dezincification process was developed with help of a high transfer degree of mass that inhibited a protective film formation of corrosive products, especially the Cu2 O film [14]. The effect of Cl ions in solutions on copper oxidation acceleration was in relation to the defect of changed Cu2 O-film structure
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2795
that had a lower protection degree on it. It assumed that the Cl ion from solution replaced some O2 ions in the oxide grid, where the primary process of selective brass dissolution (zinc dissolution) was expressed by a sudden increase of the current density, short over corrosion potential. The brass dezincification process in solution with a content of chloride ions was developed by a selective dissolution of a more reactive component (zinc), where surface diffusion of a more stable component (copper) was added. A. F. Lucey has pointed out the importance of the formation the Cu(I)-complex ion in a NaCl solution for the process of selective brass dissolution [14]. Based on corrosion testing of the copper-based alloys in sea water, de Sanchez and Schiffrin [14,15] have assumed that the formation of protective films depends a on diffusion balance of Cu(I)-chloro-complex that is formed by a primary anode reaction: Cu þ 2Cl ! CuCl 2 þe
ð1Þ
and by deposition reaction: 2CuCl 2 þ 2OH ! Cu2 O þ H2 O þ 4Cl
ð2Þ
A distinct dissolubility of Cu(I)-chloro-complex, in high salinity solutions, sets up the conditions for the development process of selective brass dissolution in the sea. In this work, the authors have tried to explain the corrosion behaviour of samples from cold-deformed CuZn-42 brass in a function of concentration with the chloride and copper(II) ions, pH-values of solutions, and the deformation degree.
2. Experimental part The samples of tested brass that were laboratory produced had the following chemical content: Cu 57.95% (99.997% purity), Zn 41.91% (99.85% purity), and others 0.14%. Samples were deformed to the following deformation degrees: 20%, 40%, 60% and 80% and sealed in cold-polymerized acrylate. Deformations were carried out in laboratory conditions by the use of cold rolling the samples between two parallel rollers, where the deformation was extended over the whole length of the sample height in all directions. A non-deformed copper electrode (99.999% purity) was used as a comparative sample. The samples for the electrochemical measurements had a constant area of P ¼ 0:38 cm2 . Before every polarization measurement, the samples were polished on emery paper (fineness #1000) and alumina rinsed with distilled water and ethyl alcohol. A saturated calomel electrode (SCE) and platinum wire were used as a reference, i.e. counter electrode, separately, in a classic threeelectrode electrochemical cell. The all potential values were given regarding the saturated calomel electrode. Tested solutions were: 101 M Na2 SO4 , 101 M Na2 SO4 + 1 · 103 M CuSO4 , 101 M Na2 SO4 + 5 · 103 M CuSO4 , 101 M Na2 SO4 + 1 · 102 M CuSO4 , 101 M Na2 SO4 + 5 · 102 M CuSO4 , 101 M Na2 SO4 + 5 · 104 M NaCl, 101 M Na2 SO4 + 5 · 103 M NaCl, 101 M Na2 SO4 + 5 · 102 M NaCl, 101 M
2796
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
Na2 SO4 + 101 M NaCl, 101 M Na2 SO4 + 5 · 101 M NaCl, 101 M Na2 SO4 + 1.0 M NaCl and preparations of high pure chemicals with distilled water. All experiments have been done with a solution volume of 50 cm3 . The pH-value 2 of solutions was adjusted with a 101 M solution of H2 SO4 . The operating temperature was 20 C, with the use of open atmosphere. Polarization measurements were carried out from a potential of open circuit to a potential of 1100 mV (SCE), with the polarization rate of 10 mV/s, by use of the linear polarization method. Corrosion current densities were determined by approximating the right part of anode Tafel’s curves to the section with the corrosion potential. The potential region from 0 (vs. SCE) to 200 mV (vs. SCE) was taken for the calculation area. An AMEL apparatus was used as well as: potentiostat (Model 553), program functional generator (Model 568), interface (Model 560/A/ log), and digital x/y-recorder (Model 863).
3. Experimental results and discussion 3.1. Influence of copper(II) ions Figs. 1 and 2 present the dependence of current density from the potential for tested copper and brass samples in an 101 M solution of Na2 SO4 with the addition of 5 · 102 M Cu2þ ions at pH 2. Corrosion potentials were established upon 15–20 min intervals, where the potential for copper electrode was shifted to more negative values, and, for brass electrode to more positive values with time. In the 101 M solution of Na2 SO4 with and without the addition of Cu2þ ions, the values of corrosion potentials for brass electrodes were approximate and lower than the values for copper electrode where a small pH-value of the solution shifted values of corrosion potential in the more negative region, that could be explained by local reactions: 1000 Cu (Ecorr=-30mV)(jcorr=0.511mA/cm2) CuZn-42(0%),(-88mV)(0.681) CuZn-42(20%),(-85mV)(0.738) CuZn-42(40%),(-85mV)(0.767) CuZn-42(60%),(-88mV)(0.912) CuZn-42(80%),(-84mV)(0.668)
E (mV/SCE)
800
600
0.1M Na2SO4 pH=2
400
200
0 2.5
3.0
3.5
4.0
4.5
5.0
log(j, µA/cm2)
Fig. 1. Polarization curves for tested samples in 101 M Na2 SO4 , at pH 2.
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2797
1000 C u , ( E corr= 3 1 m V ) ( j corr= 0 . 9 4 7 m A / c m 2) CuZn-42(0%),(-16mV)(1.238) CuZn-42(20%),(10mV)(1.333) CuZn-42(40%),(10mV)(1.378) CuZn-42(60%),(-10mV)(1.448) CuZn-42(80%),(10mV)(1.211)
E (mV/SCE)
800
600
0.1M Na2SO4 + 5*10-2M Cu2+ pH=2
400
200
0 2.8
3.2
3.6
4.0
4.4
4.8
5.2
log( j, µA/cm2)
Fig. 2. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 102 M Cu2þ , at pH 2.
Zn ! Zn2þ þ 2e
ð3Þ
Cu2þ þ 2e ! Cu
ð4Þ
on sample surface [16]. The concentration increase of Cu2þ ions resulted in the movement of corrosion potentials into a less active region, pointing out that the balance realization between Cu2þ /Cuþ ions on a boundary of metal–solution and the values of current densities are also increased [13]. The lowest values of corrosion current densities (jcorr ) were for brass with a deformation degree of 80%. The dezincification process, present here, is a cause of the increase of the density values of corrosive currents as well as the non-existence of a protective film that mainly consisted of Cu2 O on a sample surface at pH 2. At higher pH-values, protective film, besides Cu2 O and CuO oxides, also consisted of a Cu3 (SO4 )2 Æ 4H2 O component that was stable in a wide range of pH-values [3,17]. The corrosive morphology of cold-deformed samples has a disadvantage in acid solutions because it has a resistant protective film only to a certain degree, under conditions of local destruction (dissolution) and gives corrosive openings that a little bit later resulted in spreading of one type of corrosion [13]. Cuþ ion, present in Cuþ compounds, probably is an intermediate form in a reaction of dissolution followed by a controlled oxidation rate of Cuþ to Cu2þ ions, where more favourable conditions have been provided for brass dezincification [3,14,17]. 3.2. Influence of chloride ions The polarization curves in Figs. 3 and 4 are shown for copper and brass with various deformation degrees in a 101 M Na2 SO4 solution with the addition of chloride ions in a concentration of 5 · 104 and 5 · 103 M at pH 2. Immediately
2798
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
C u , ( E corr= - 5 5 m V ) ( j corr= 0 . 7 2 7 m A / c m 2) CuZn-42(0%),(-149mV)(1.04) CuZn-42(20%),(-171mV)(1.108) CuZn-42(40%),(-136mV)(1.166) CuZn-42(60%),(-148mV)(1.42) CuZn-42(80%),(-170mV)(1.035)
1000
E (mV/SCE)
800
600
0.1M Na2SO4 + 5*10-4 ClpH=2
400
200
0 2.0
2.5
3.0
3.5
4.0
4.5
5.0
log( j, µA/cm2)
Fig. 3. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 104 M Cl , at pH 2.
1200 C u ( E corr= - 6 1 m V ) , ( j corr= 0 . 9 8 8 m A / c m 2) CuZn-42(0%),(-195mV),(1.298) CuZn-42(20%),(-188mV),(1.585) CuZn-42(40%),(-160mV),(1.688) CuZn-42(60%),(-170mV),(1.721) CuZn-42(80%),(-220mV),(1.278)
1000
E (mV/SCE)
800 600
-
0.1M Na2SO4 + 5*10-3M Cl pH=2
400 200 0 -200 2.5
3.0
3.5
4.0
4.5
5.0
log( j, µA/cm2)
Fig. 4. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 103 M Cl , at pH 2.
upon the setting of corrosion potential, all tested samples are moved into an active state where they are in the whole range of tested potential. With an increased Cl ion concentration, the corrosion potentials have a tendency of movement into a more active region for all the tested electrodes, inducing a lower degree of anode process control. The most positive values for Ecorr in the Cu electrode are in all tested solutions. The anode polarization curves have a similar form unless the values for jcorr are also increased with an increased brass deformation degree up to 60%, whereas values of the corrosion current densities are the lowest for a deformation degree of 80%. Only copper electrode has lower values for jcorr of brass with a deformation degree of 80%.
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2799
Densities of corrosion currents are higher than solutions with a concentration of chloride ions of 5 · 103 M. It could be noted here that the values for jcorr with a 101 M solution of Na2 SO4 and an addition of chloride ions in a concentration of 5 · 103 M are higher than values for jcorr with a solution concentration of Cl ions of 5 · 104 and 5 · 102 M, even higher than some values with a solution concentration of chloride ions of 101 M. Figs. 5 and 6 give anode polarization curves for tested brass samples in a 101 M solution of Na2 SO4 with an addition of chloride ions in a concentration of 5 · 102 and 1.0 M. The values of corrosive potentials relevant to an 101 M solution of Na2 SO4 , with an addition of chloride ions in concentration of 5 · 104 M, were shifted into a more active region. The values of corrosion current densities were moved into a more positive area at small pH-values of solution and an increase of brass deformation
C u , ( E corr= - 1 4 5 m V ) ( j corr= 0 . 6 7 8 m A / c m 2) CuZn-42(0%),(-232mV)(1.083) CuZn-42(20%),(-226mV)(1.122) CuZn-42(40%),(-240mV)(1.182) CuZn-42(60%),(-218mV)(1.192) CuZn-42(80%),(-225mV)(1.008)
1000
E (mV/SCE)
800 600
0.1M Na2SO4 + 5*10-2M Cl pH=2
-
400 200 0 -200 2.5
3.0
3.5
4.0
4.5
5.0
log( j, µA/cm2)
Fig. 5. Polarization curves for tested samples in 101 M Na2 SO4 + 5 · 102 M Cl , at pH 2.
C u , ( E corr= - 2 7 3 m V ) , ( j corr= 1 . 4 4 2 m A / c m 2) CuZn-42(0%),(-335mV)(3.158) CuZn-42(20%),(-330mV)(3.492) CuZn-42(40%),(-330mV)(3.752) CuZn-42(60%),(-330mV)(4.11) CuZn-42(80%),(-333mV)(3.005)
1000
E (mV/SCE)
800 600
-
400
0.1M Na2SO4 + 1.0M Cl pH=2
200 0 -200 2.5
3.0
3.5
4.0
4.5
5.0
5.5
log(j, µA/cm2)
Fig. 6. Polarization curves for tested samples in 101 M Na2 SO4 + 1.0 M Cl , at pH 2.
2800
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
degree up to 60%, where values for jcorr are the highest. The most negative values for jcorr are in brass with a deformation degree of 80%. On anode polarization curves, the presence of anode peaks were noticed, presenting an active dissolution of tested samples and formation of Cu(I)-chloro complex formed by a primary reaction: Cu + 2Cl $ CuCl 2 + e . The Cu(I)-chloro 2 complex that will be formed (CuCl2 or CuCl3 ) depends on the concentration of chloride ions. A distinct dissolution of Cu(I)-chloro-complex, in solutions with an increased concentration of Cl ions, makes conditions for development of the selective dissolution process of brass (dezincification) as well as for a loss of the already formed copper-chloride film on its surface. In concentration values of chloride ions of 5 · 102 M, a drop of value for jcorr was noticed regarding to the solutions with an addition of chloride ions in a concentration of 5 · 103 M, explained by an inhibitory effect of higher concentrations of chloride ions (see Fig. 7). Anode polarization curves show that the lowest values for corrosive current densities are in Cu electrodes; and, from brass samples, that brass with a deformation degree of 80% has the lowest values for jcorr . Brass with a deformation of 60% has the highest values for jcorr . Regarding the Cu electrode, the anode peaks of brass electrodes are moved to more positive potential values. Flow of anode polarization curves pointed out that anode dissolution of brass in solutions with chloride ions started with an initial primary dissolution of zinc, and then zinc and copper, depending on potential region, but not in a stoichiometric relationship. The primary process of selective brass dissolution was expressed by a sudden increase of current density immediately upon the corrosion potential, at the beginning of anode polarization. By chemical analysis of the analyzed solution, presence of some Zn2þ and Cu2þ /Cuþ ions was found to point out that, under the afore mentioned experimental condition, the dezincification process of brass could be developed with the following process of its anode dissolution. Various electrode potentials of zinc, that is copper, have considerable influence on the degree, intensity, and rate of the pri-
4.0 3.5
CuZn-42 (60%) pH-2 0.1M Na2SO4
jcorr (mA/cm2)
3.0 2.5 2.0 1.5 1.0 0.0
0.2
0.4 0.6 CCl- (M)
0.8
1.0
Fig. 7. Dependence of jcorr on concentration of Cl ions for CuZn-42 (60%), at pH 2.
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
2801
mary anode dissolution. Zinc passage into a solution in the form of Zn2þ ion and the formation of complex and undissolved salts contributes to the irreversible nature of development in the dezincification process under the above mentioned conditions. Metallic copper drops back, partly as a porous metallic mass on an electrode surface of CuZn-42 brass, whereas the other part of Cu ions diffuses on a brass sample surface, where it is separated again in a form of copper and various Cu compounds by further kinetics control of total alloy dissolution. According to the balance diagram for content of Cu–CuCl , at various pH-values of NaCl solution, it was found that the major participation of more dissoluble 2 copper compounds (Cuþ , CuCl 2 , CuCl3 ) are present at higher pH-values relevant to the participation of less dissoluble copper compounds (CuCl, Cu2 O) [14]. Based on the obtained results, the brass dezincification process is developed, in solutions with a content of chloride ions, by the selective dissolution of a more reactive component (zinc) where surface diffusion of more a stable component (copper) has been added to. The dezincification process was developed by help of a high degree of mass transfer that inhibited the protective film formation of corrosive products, especially Cu2 O film [8,15,16]. Lesser values of corrosion current densities, in tested samples with a deformation degree of 80% and at higher concentrations of chloride ions, are explained by an easier process of substitution O2 ions with chloride ions in crystal lattice. Also, a protective film is formed faster in deformed samples of 80% on its surface that causes lower values for jcorr .
4. Conclusion Based on the presented testing results of corrosion behaviour of CuZn-42 brass, with various deformation degrees, the following conclusions could be made: 1. Increase of Cu2þ ion concentration results in an increased value for jcorr . 2. Inhibiting effect of chloride ions was noticed for concentrations of ions of 5 · 102 and 101 M in an 101 M solution of Na2 SO4 . 3. Increase of a concentration of the Cl ions, except for values of 5 · 102 and 101 M, resulted in a significant increase of value for jcorr . 4. A pH-value of solution resulted in an increase of value for jcorr . 5. The process of dezincification and anode dissolution of CuZn-42 brass was developed in the whole range of tested potentials, and pointed out the unstability of formed films on their surface. 6. The complex form of anode curves pointed out the phase reactions of selective a dissolution process of CuZn-42 brass. 7. Surface film formation of corrosion products was developed through some phases of and distinctive presence of metallic copper layers and spot corrosion present on surface of corroded electrode. 8. Brass with a deformation degree of 80% has the lowest values for corrosion current densities of brass.
2802
Z. Avramovic, M. Antonijevic / Corrosion Science 46 (2004) 2793–2802
Acknowledgements The authors wish to thank the Ministry of Science and Technology of Serbia for financial support (No. 1622).
References [1] R.H. Heidersbach Jr., Ellis D. Verink Jr., Corrosion-NACE 28 (11) (1972) 397. [2] T.K.G. Namboodhiri, R.S. Chaudhary, B. Prakash, M.K. Agrawal, Corrosion Science 22 (1982) 1037. [3] R.K. Dinnappa, S.M. Mayanna, Corrosion Science 27 (1987) 349. [4] R.B. Rebak, J.R. Galvele, Corrosion Science 29 (1989) 1003. [5] L. Burzynska, J. Stoch, Z. Zembura, Solid State Ionics 38 (1990) 179. [6] L. Burzynska, Z. Zembura, Polish Journal of Chemistry 66 (1992) 503. [7] L. Burzynska, A. Maraszewska, Z. Zembura, Corrosion Science 38 (1996) 337. [8] Lj. Krstulovic, B. Kulusic, Chemistry and Industry 45 (5) (1996) 177. [9] T.P. Hoar, C.J.L. Booker, Corrosion Science 5 (1965) 821. [10] S. Torchio, F. Mazza, Corrosion Science 26 (1986) 813. [11] S.S. Birley, Desmond Tromans, Corrosion-NACE 27 (7) (1971) 297. [12] R.C. Newman, G.T. Burstein, Corrosion Science 21 (1981) 119. [13] S. Torchio, Corrosion Science 26 (1986) 133. [14] B. Kulusic, Lj. Krstulovic, J. Ivic, Chemistry and Industry 39 (7) (1990) 315. [15] S.R. De Sanchez, D.J. Schiffrin, Corrosion Science 28 (1988) 141. [16] F. Mansfeld, G. Liu, H. Xiao, C.H. Tsai, B.J. Little, Corrosion Science 36 (12) (1994) 2063. [17] A.M. Beccaria, E.D. Mor, G. Poggi, F. Mazza, Corrosion Science 27 (1987) 363.